Anemia is a frequent complication of chronic kidney disease (CKD) and is associated with a decrease in the quality of life of patients, as well as an increase in morbidity and mortality and CKD progression.1,2 Although it has been classically described that the main cause of anemia in CKD is the inadequate production of endogenous erythropoietin, in recent years other contributing factors have been recognized, such as a decreased erythropoietic response of the bone marrow due to uremic toxins and inflammatory state, iron deficiency and decreased iron availability for erythropoiesis due to increased hepcidin levels, shortened red blood cell survival or vitamin deficiencies (vitamin B12 or folic acid), among others.3

Iron deficiency (absolute or functional) is common in CKD patients who do not receive dialysis (non-D CKD)4–7 and can cause anemia and hyporesponsiveness to erythropoiesis-stimulating agents (ESAs). Furthermore, it has recently been observed that iron deficiency per se in this population has been associated with a higher risk of adverse outcomes.7–9 Iron administration can increase hemoglobin levels (Hb) in patients with CKD and anemia, and even in some patients with non dialysis CKD, allows to achieve target levels of Hb without ESA.10 If the patient receives ESA, adequate iron parameters must also be ensured before and during treatment, in order to achieve an adequate response and reduce the doses of ESA. However, although the risk-benefit balance is favorable, it should not be forgotten that iron therapy presents some adverse effects and potential risks, the evidence of which will be the subject of the present discussion.

The clinical guidelines for the treatment of anemia in Nephrology advise correction of iron deficiency in the presence of anemia.11–13 However, in other diseases, such as heart failure and in other situations, a clinical benefit of iron deficiency correction has been demonstrated, regardless of the presence or absence of anemia,14 so correction of iron deficiency per se in CKD should be an area of ​​future research in Nephrology.

Classically, the assessment of the iron status in the body in CKD patients has been based mainly on 2° parameters: ferritin levels and the transferrin saturation index (TSAT). However, these values, although widely available, are easily influenced by different factors. This limitation could be improved by using other markers (less accessible and more expensive), such as the percentage of hypochromic red blood cells (provided the sample is processed within 6 h from its extraction), the reticulocyte Hb content, or hepcidin, 15–18 although only the first 2  are recommended as alternatives in the different current clinical guidelines. For this reason, and given the lack of reliability of new markers, we must continue to use the classic markers.

There is consensus on the definition of iron deficiency and the different subtypes that compose it in CKD15,16:

• a

  Absolute iron deficiency: situation depletion of iron stores, defined as a serum ferritin concentration < 100 ng/ml and TSAT < 20%.15–17

• b

  Functional iron deficiency: defined as an TSAT < 20%and normal/high ferritin levels. In functional iron deficiency, iron stores are normal or high, but iron cannot be adequately mobilized from the reticulo-endothelial system. The causes of functional iron deficiency are treatment with ESA (the iron needs by the bone marrow exceed the ability to mobilize iron from the stores) and inflammation, mainly. In this case, the indication for iron therapy is controversial in terms of the target levels of ferritin and TSAT to achieve in order to obtain a clinical benefit for the patient.15,16

There is unanimity in the clinical guidelines for the indication of iron therapy in patients with CKD, anemia and iron deficiency if there is an absolute iron deficiency (ferritin < 100 ng/ml and TSAT < 20%), but there is no unanimity in other situations, it is especially in relation to the administration of iron in patients with functional iron deficiency.11,19 All of them coincide in correcting iron deficiency with the aim of improving quality of life, avoiding transfusions and delaying the need or reducing ESA doses in anemic patients with CKD and iron deficiency. Table 1 shows the indications for iron therapy according to the different guidelines. European guidelines (ERBP) adopt a more conservative approach in anemic patients with non-dialysis CKD, indicating it to patients without ESA if TSAT < 25% and ferritin < 200 ng/dl if an increase in Hb concentration is desired without start ESAs or if the patient is undergoing ESA treatment, iron therapy would be indicated when the TSAT is< 30% and ferritin < 300 ng/dl if an increase in Hb levels or a reduction in the ESA dose is desired.9 However, the KDIGO guidelines established with an evidence 2C that iron therapy can be initiated in patients with IST < 30% and ferritin < 500 ng/dl, whether or not they receive ESA.8

However, the results of the FIND-CKD studies10 in patients with non-D CKD or the PIVOTAL study in hemodialysis patients20 may modify these objectives in the next guidelines.

European guidelines also state that treat or not to CKD patients, Hb  > 12 g/dl and iron absolute deficit is an open question,12 given the absence of evidence of the risk-benefit profile.

The recommendation of clinical guidelines on the administration of iron do agree in patients with CKD anemic (Hb < 12 g/dl) with TSAT < 20% and ferritin < 100 ng/ml. No recommendations in anemic patients with non dialysis CKD and functional iron deficiency, given the absence of evidence.11

It is not appropriate to extrapolate the threshold Hb applied for starting ESA prescription (Hb < 10  g/dl), which are lower, based on the results of clinical trials with these agents.11–13 Iron, unlike ESAs, is a necessary factor for an efficient erythropoiesis, but not a growth factor, so there is no risk of overshooting or exceeding the Hb level above the target limit, since the red blood cell production will be regulated by endogenous EPO and if ESA is used, its dose can be reduced, which would improve safety with the latter and reduce costs. On the other hand, we already know from the post hoc analysis of the CHOIR and TREAT studies that the cardiovascular events (CV) of the group with a target Hb >13 g/dl are more associated with the high doses of ESA used to achieve this Hb level, than the Hb value finally reached.21,22 In this regard, patients with CKD and spontaneous Hb > 13 g/dl have no more events than those with lower levels of Hb.23

The different clinical guidelines advocate first the administration of iron to replenish the deposits and only when this has been achieved and the anemia persists they suggest starting treatment with ESA.11–13 Although the KDIGO guidelines seem to favor intravenous (IV) iron administration in patients with non-D CKD, they also consider oral iron therapy (PO) as an alternative for 1–3 months in these patients.12

On the other hand, the European recommendations (ERBP) advise that in patients with non-D CKD and mild-moderate anemia, oral iron therapy would be the first-line treatment for a minimum of 3 months.11 IV iron therapy would be indicated if:

• ­

  The objectives of iron parameters are not reached with oral iron therapy for 3 months or when there is intolerance or malabsorption of oral iron.

• ­

  Patients with severe anemia and iron deficiency in whom a rapid Hb response is required.

• ­

  Patients with chronic inflammatory processes showing functional iron deficiency (TSAT < 20% with high-normal ferritin), especially if they require ESA.

The NICE guidelines also advocate the administration of oral iron in patients with non-D CKD, although they comment that some patients will require IV iron. It should be noted that this guide indicates that when IV iron therapy is prescribed in CKD patients who are not on hemodialysis, a high-dose, low-frequency strategy is recommended as the treatment of choice,13 contrary to the strategy suggested in patients on hemodialysis.

The guidelines also recommend avoiding tissue iron accumulation, so they advise that, during treatment with iron, the limits of TSAT of 30% and a ferritin of 500 ng/ml should not be exceeded, both in patients with non-D CKD, and in CKD-5D.11

Several studies and meta-analysis have compared the efficacy and safety of IV iron vs. oral iron in patients with non-D CKD, and the most relevant ones will be mentioned here. Table 2 shows the most important randomized IV iron studies in patients with non-D CKD.

Qunibi et al. compared treatment with ferric carboxymaltose (FCM) 1 g vs. oral iron (65 mg/d). After 8 weeks of treatment, the increase in Hb levels was higher with FCM (1.31 g/dl vs. 0.83 g/dl, p < 0.01) and a higher percentage of patients than receiving FCM had an increase in Hb ≥ 1 g/dl compared to oral iron therapy (60.4% vs 34.7% of patients), regardless of whether or not they were receiving ESA treatment. Treatment with FCM alone was as effective as combined treatment with PO iron and ESA, and combined treatment with FCM and ESA was superior to treatment with PO iron and ESA.24 Similar results were reported in the PROGRESS study with Iron isomaltoside (IIS).25

The FIND-CKD study is the largest global multicenter study with long follow-up, which compared IV and PO iron treatment in patients with non-D CKD and anemia, not treated with ESA. The study included 626 patients who were randomized to receive FCM 1 g and then every 4 weeks, if necessary, to achieve ferritin levels between 400−600 ng/ml, or FCM 200 mg and then every 4 weeks if it was necessary to maintain ferritin levels between 100−200 ng/ml or iron PO (ferrous sulfate at a dose of 200 mg/day of elemental iron). The primary endpoint was time to initiate an alternative treatment for anemia or occurrence of a trigger of Hb, as defined 2 consecutive values Hb < 10 g/dl or after week 8 without an increase ≥ 0.5 g/dl between consecutive values. 76% of the patients maintained Hb ≥ 10 g/dl or required additional treatment for anemia when treated with FCM in order to achieve high levels of serum ferritin (hazard ratio [HR] 0.65, IC 95%: 0.44 to 0.95, p = 0.026 for FCM and high ferritin vs. oral iron). Likewise, the percentage of patients who achieved an increase in Hb ≥ 1 g/dl was higher in the high FCM ferritin group (56.9% vs. 34.2% low FCM-ferritin group and 32.1% oral iron, p < 0.01). The high FCM-ferritin group had higher Hb and ferritin levels than the other 2 groups throughout the 52 weeks of study. Regarding safety, there were no differences between the groups in terms of adverse events or serious adverse effects between the different groups, but there was a higher dropout rate in the PO iron group, especially due to gastrointestinal adverse events.10

The REVOKE study was a single-center, randomized clinical trial that compared the effect of IV iron sucrose (IS) vs. oral iron therapy in patients with non-D CKD on renal function as the primary event.26 The study was stopped prematurely due to the increased risk of adverse events in the group treated with IV iron, which generated controversy in the nephrology community. However, the number of patients who experienced adverse events was similar in both groups (55 vs. 58%), but when the total number of adverse events was considered, it was higher in the IV iron group: 199 per 100 patient-years vs. 168.4 per 100 patient-years. Serious adverse effects occurred repeatedly in the same patients in the IV iron group (201 events in 37 patients vs 176 events in 40 patients in the PO iron group), but this difference only reached significant differences after adjusting for baseline characteristics . Thus, it is a single-center study in which only 99 patients completed the study. Most adverse events occurred long after the intervention was completed, which ended 8 weeks after randomization, so adverse effects that occurred after 3 months may or may not be related to treatment. The safety signal occurred due to the presence of repeated adverse effects in the same patients and after adjusting. Therefore, and according to the Cochrane criteria (version 5.1.0.) this study has a high risk of bias27 (Table 2).

The meta-analysis by Shepshelovich et al. included 24 trials, 13 involving 2,369 patients with CKD stages 3–5 and 11 that included 818 patients with CKD stage 5D. Patients treated with iron IV were more likely to get an increase in Hb > 1 g/dl (HR 1.61, 95% CI: 1.39–1.87, p < 0.00001) and ferritin levels were significantly higher in all IV iron groups vs. the iron PO group (mean difference between formulations 238.9 ng/ml, 95% CI: 194.3, 283.5, p < 0.00001). Regarding safety, analysis showed comparable rates of adverse and serious adverse events from iron IV and PO (1.06, 95% CI: 0.88–1.28, p = 0.53). However, the follow-up periods were generally limited to 3 months, thus the authors acknowledge that more trials with longer follow-up periods are required to understand the consequences of different treatments on Hb levels and clinical outcomes. The authors concluded that the IV iron treatment is preferred, both for patients with CKD stage 5D, as for those with nonD non D CKD stages 3-5.28 These results have recently been confirmed by the meta-analysis carried out by the Cochrane in which 39 studies (3,852 participants) were examined that compared IV iron supplements versus PO.29

Finally, in a post hoc analysis of the FIND-CKD study, it was described that in those patients randomized to oral iron therapy, at 4 weeks after starting treatment, only 21.6% of anemic patients with non-D CKD and iron deficiency showed an increase in Hb ≥ 1 g/dl. Among those who did not respond at 4 weeks, less than 30% responded at the end of the study (week 52), suggesting that the lack of early response to oral iron treatment could be a reason to consider IV iron therapy,30 which would make it possible to indicate IV iron therapy earlier and correct anemia earlier.

The REPAIR-IDA study analyzed a high-dose, low-frequency strategy with a low-dose, high-frequency strategy. Study31 included 2,584 patients who were randomized to 2doses of FCM 750 mg in 1 week or IS 200 mg administered in up to 5 infusions in 14 days. In the FCM group, a greater number of patients achieved an increase in Hb ≥ 1 g/dl (48.6% vs. 41.0%; 95% CI: 3.6%–11.6%). There were no differences in the composite safety event of cardiovascular death, heart attack and stroke, but there were more episodes of transient hypertension in the FCM group. Therefore, it is concluded that this high-dose, low-frequency strategy is as safe and effective as a multiple-infusion strategy of low-dose ISa.

The FERWON-NEPHRO study was a randomized, open-label, multicenter trial conducted in the USA. Patients were randomized 2: 1 to a single dose of 1,000 mg of IIS 1,000 or IS in doses of 200 mg up to 5 times. In a period of 2 weeks. A total of 1,538 patients were included (mean eGFR 35.5 ml/min/1.73 m2). The safety objective was met based on the absence of significant differences in the incidence of severe or severe hypersensitivity reactions in the IIS and IS groups (0.3% vs. 0%; risk difference: 0.29% [95% CI: –0.19, 0.77; p > 0.05]). The incidence of the composite CV adverse events was significantly lower in the ISS group vs. IS (4.1% vs. 6.9%; p = 0.025, of which heart failure occurred in 0.7% in IIS vs. 2.2% in IS, p = 0.02; hypertension 1, 1% vs. 2%, p = 0.16, and atrial fibrillation 0.3% vs. 1.2%, p = 0.067). Compared with the IS, the IIS produced a higher Hb increase during the first 4 weeks (1.06 vs. 0.91 g/dl, p ≤ 0.021) and the change in Hb at 8 weeks showed no inferiority. The authors concluded that, compared to multiple doses of IS, a single dose of IIS induced a non inferior hematologic response at 8 weeks, comparatively low rates of hypersensitivity reactions, and a significantly lower incidence of the composite CV adverse event.32

Thus, the high-dose, low-frequency strategy recommended by the NICE guidelines13 demonstrates better convenience for the patient with a good efficacy and safety profile.

Table 3 shows the benefits as well as the limitations and risks of both PO iron and IV iron. Oral iron supplements can be taken as capsules, tablets, chewable tablets, and in liquid solution. This route of administration is simple, cheaper and preserves the venous capita, which should be considered in these patients given the future need for hemodialysis treatment. In patients with non-D CKD or on peritoneal dialysis in which oral iron is prescribed, the doses recommended in the guidelines in an adult patient will be around 200 mg/day of elemental iron divided into 2–3 doses (better ferrous salts for its better absorption) and preferably on an empty stomach. However, recent studies in non-renal patients seem to show greater efficacy and tolerance of low doses of oral iron, since small increases in iron levels increase hepcidin levels, limiting its intestinal absorption. Recent studies show better absorption of oral iron when administered every other day or in a single daily dose,33,34 which suggests that changing the frequency of administration increases the efficacy of treatment in iron-deficient patients and may improve tolerability. The main problems associated with oral iron treatment in CKD are gastrointestinal intolerance, intestinal absorption problems, or lack of therapeutic compliance associated with its adverse events. Although there are differences in the appearance of side effects, constipation and diarrhea, along with the change in the color of the stool (black) are very common. Nausea and vomiting can occur, and liquid iron formulations can stain teeth. Since non-heme iron absorption is modest, high doses of oral iron can induce oxygen-free radical-mediated toxicity from unabsorbed iron on the intestinal mucosa. On the other hand, oral iron supplements could negatively affect the intestinal microbiota, already altered in CKD.35 This phenomenon has not been sufficiently studied in kidney patients, but it could negatively affect the production of uremic toxins or increase intestinal permeability to the passage of bacterial products into the circulation, aggravating inflammation in these patients,36 as has been shown in other diseases.37

It has been suggested that alternative oral iron compounds might be more effective or better tolerated in this population (sucrosomial iron, ferric citrate, polypeptide heme iron, or ferric maltol, etc.). Of these, ferric citrate has shown repletion of iron stores, partial correction of anemia, and also a reduction in phosphorus, PTH-i and FGF-23.38 Ferric citrate has been approved as a phosphate binder in dialysis patients and has recently also been approved by the Food and Drug Administration for the treatment of iron deficiency anemia in patients with CKD. At the present time, it is not available in Spain (https://www.accessdata.fda.gov/drugsatfda_docs/label/2017/205874s013lbl.pdf). However, the evidence in CKD of the other compounds is limited to recommend their use and in some cases they are approved as nutritional supplements and not as drugs, with limitations in terms of evidence of efficacy and safety, in addition to having a higher cost for the patient.39,40

In the administration of IV iron in patients with non-D CKD, the preservation of the venous capital should be considered due to the future need for vascular access for hemodialysis. For this, it is convenient to administer iron coinciding with the performance of analytical controls, thus avoiding more punctures, and the administration of iron at high doses and with less periodicity, as recommended by the NICE guidelines.13 The most important adverse events related to the administration of IV iron compounds are hypersensitivity reactions to it or to any of the excipients. In the aforementioned meta-analysis, a higher risk of arterial hypotension was also reported with IV administration vs. PO.28,29 Likewise, the development of transient hypophosphataemia has been described with IV iron therapy, especially with FCM.41,42 Although it has been described in some studies, and in relation to the administration of IV iron, increased oxidative stress, impaired kidney function, risk of infections or iron accumulation, this has not been confirmed by others. In this sense, the aforementioned meta-analysis did not observe differences in mortality, infection, changes in glomerular filtration rate, or the need for dialysis between IV and oral iron therapy.28,29 More recently, in the PIVOTAL study in hemodialysis patients, those patients randomized to high doses of IV iron did not demonstrate an increased risk of infections.43 These concerns led to a controversial document from the KDIGO in which 4 of these aspects were analyzed and concluded44:

• ­

  Hypersensitivity reactions: The risk of hypersensitivity reactions secondary to IV iron is currently low. However, recent recommendations from the European Medicines Agency (EMA) have sensitized the medical community to these possible negative consequences of serious reactions that occur in circumstances in which there is no adequate organization for such urgent care.

• ­

  Inflammation and oxidative stress: IV iron can cause oxidative damage to the DNA of peripheral blood lymphocytes in hemodialysis patients, protein oxidation and lipid oxidation. Although high doses of IV iron are associated in some studies with a higher mortality rate in the United States dialysis population, it is not clear that iron administration promotes atherosclerosis, arterial remodeling or CV disease, being controversial and inconclusive the results of all studies so far. In fact, at the last EDTA congress a prespecified secondary analysis of the PIVOTAL study was presented, which demonstrated that the proactive iron strategy reduced the risk of type 1 myocardial infarction (of atherosclerotic origin) (European Renal Association-European Dialysis and Transplant Association [ERA-EDTA] 57th Congress. Abstract MO016. Presented June 8, 2020).

• ­

  IV iron and infections: Despite inconclusive results from systematic reviews and meta-analyzes, the benefits and risks of iron therapy should be considered and the administration of IV iron should be avoided in the presence of active infection. However, it must be taken into account that the suspension of iron therapy can lead to an iron deficiency, which in turn induces an increased risk of infection.

• ­

  Iron overload: Pathological iron overload represents a condition of high body iron content associated with signs of organ dysfunction that are presumably caused by excess iron. Although some studies have shown an increase in iron content at the liver level, its clinical relevance is unknown and deposits have not been observed in other territories, such as at the cardiac or pancreatic level.45–47 However, an association between hepatic iron overload and hepatic steatosis has recently been described,48 whose clinical significance is unknown. Despite the lack of evidence, the document recommends avoiding the combination of elevated TSAT and ferritin values.

These conclusions seem to be confirmed in a recent meta-analysis of clinical trials and observational studies in hemodialysis (receiving higher doses of IV iron) that has not found an increased risk of mortality, infections, hospitalization or CV disease in patients who received high doses of IV iron However, this meta-analysis has its limitations, as the follow-up was relatively short and the studies were not designed or powered to analyze directly measurable and clinically relevant outcomes . In observational studies there was a high degree of heterogeneity and inevitably presented a prescription bias.49 On the other hand, a recent review of recommended reading has not shown that a positive iron balance in CKD is associated with clinical toxicity.50 The aforementioned results of the PIVOTAL study also demonstrate the CV and infection risk safety of a proactive IV iron strategy in hemodialysis patients.20,43

• ­

  Clinical relevance of iron deficiency, regardless of anemia in CKD, evidence from observational studies points to a worse prognosis associated with iron deficiency in this population.

• ­

  Long-term efficacy and safety of oral and IV iron therapy in patients with CKD.

• ­

  Effect of oral iron therapy on the intestinal microbiota, intestinal permeability and systemic inflammation in CKD.

• ­

  More effective dose and administration intervals (daily or every other day) of oral iron therapy in CKD.

• ­

  Controlled studies evaluating the impact of iron deficiency correction in the absence of anemia on morbidity and mortality, quality of life and physical capacity.

The authors have not received support for the submitted work.

AC reports grant and honoraria for lectures and consulting from Vifor Pharma, grant and honoraria for consulting from AB-Biotic, honoraria for lectures and consulting from Astellas, Astra Zeneca and Novo Nordisk; honoraria for consulting from GSK and honoraria for lectures from Amgen and from Bayer, outside the submitted work.

MJP and JLG report honoraria for lectures from Vifor Pharma, outside the submitted work.

PS reports honoraria for lectures from Vifor Pharma, and a grant and participation on an Advisory Board from Astellas, outside the submitted work.

BQ reports honoraria for lectures or support for attending meetings and/or travel from Vifor-Pharma, Astellas, Amgen, Ferrer, Novartis, AstraZeneca, Sandoz, Laboratorios Bial, Esteve, Sanofi-Genzyme, Otsuka; participation on Advisory Boards from AstraZeneca and Laboratorios Bial, outside the submitted work. Currently he is the Secretary of Sociedad Española de Nefrología (S.E.N.).

LMR reports honoraria for lectures from Vifor Pharma and Astellas, and support for attending meetings and/or travel from Baxter.

JP reports participation as Principal Researcher in clinical trials on anaemia conducted by Amgen, Roche, Astellas and GSK, from Amgen, Vifor-Pharma, GSK and Astellas, and honoraria as advisor from Astellas and GSK.